DEEP LEARNING FRAMEWORK FOR DIAGNOSTICS AND PATIENT-SPECIFIC DESIGN - PowerPoint PPT Presentation

DEEP LEARNING FRAMEWORK FOR DIAGNOSTICS AND PATIENT-SPECIFIC DESIGN OF BIOPROSTHETIC HEART VALVES ADITY TYA A BALU SAHI HITI TI NALL LLAGOND GONDA MING NG-CHEN HEN HSU SOUM UMIK SARKAR RKAR ADAR ARSH SH KRISH SHNAM AMUR URTH THY March 18, 2019 1

Heart Diseases • Leading cause of death $6.336B In both the US and the • world 1 in every 4 deaths • A heart attack every 40s • $11.588B • Loss of revenue $200 billion each year • $6.116B $3.518B March 18, 2019 2

Valvular Diseases • Valvular Heart diseases • Affects more than 2.5% of US population • Causes • Calcification (Narrowing at the opening) • Regurgitation (Leakage and reverse flow) • Intervention • Surgical replacement • 90,000 prosthetic heart valves per year [1] https://www.webmd.com/heart-disease/guide/heart-valve-disease#1 March 18, 2019 3

Artificial Heart Valves • Mechanical Valve • Advantages • Durable • Disadvantages • Causes damage to blood cells • Need blood thinner to prevent clots • Noisy (can cause sleepless nights) • Bioprosthetic Valves Suture Ring • Use bovine or porcine pericardium • Advantages • Replicates the valve tissue • Disadvantages • Durability due to fatigue Mechanical Bioprosthetic • Prone to calcification valve valve March 18, 2019 4

Patient-Specific Replacement Heart Valves • Common sizes • Disadvantages of wrong sizing • Poor valve function (regurgitation, low flow rate) • Durability • BHV Replacements • 10 years [2] https://www.medtronic.com/ca-en/healthcare-professionals/products/cardiovascular/heart-valves-surgical/mosaic-mosaic-ultra-bioprostheses.html [3] https://www.heartvalvechoice.com/tissue-vs-mechanical-heart-valve/ March 18, 2019 5

Patient-Specific Design of Heart Valves • Design heart valves for every patient using their medical results An aortic bioprosthetic heart valve with A view of one leaflet of the heart valve its placement on aorta with its parametric curve boundary March 18, 2019 6

Valve Function • Coaptation Area • Open area Coaptation Area Open Area March 18, 2019 7

Design of BHV • Custom design requires evaluation of the valve function • Simulation speeds up the process A view of one leaflet of the heart valve with its parametric curve boundary March 18, 2019 8

Simulations of BHVs • Imaging analysis for surgical decision making is difficult • Simulation of physics is necessary Phase contrast MRI image data [4] M. C. Hsu et al., “Dynamic and fluid – structure interaction simulations of bioprosthetic heart valves using parametric design with T-splines and Fung-type material models,” Computational Mechanics, 55 (2015) 1211-1225 March 18, 2019 9

Outline Computational ML Framework for Data Representation Results and Modeling of BHVs Valve Mechanics and Data Generation Conclusions March 18, 2019 10

Computational Modeling Frameworks for BHVs • Reconstruct the heart valve from medical images • Generate geometric representation of the heart valve (NURBS) • Perform valve closure simulations • Use Isogeometric analysis [5] S Morganti, F Auricchio, DJ Benson, FI Gambarin, S Hartmann, TJR Hughes, and A Reali. Patient-specific isogeometric structural analysis of aortic valve closure. Computer Methods in Applied Mechanics and Engineering, 284:508 – 520, 2015 [6] Fei Xu, Simone Morganti, Rana Zakerzadeh, David Kamensky, Ferdinando Auricchio, Alessandro Reali, Thomas JR Hughes, Michael S Sacks, and Ming-Chen Hsu. A framework for designing patient-specific bioprosthetic heart valves using immersogeometric fluid – structure interaction analysis. International journal for numerical methods in biomedical engineering, 34(4):e2938, 2018. March 18, 2019 11

Reconstruction of Aortic Valve Reconstruction of Aortic Root from CTA for a patient [5] S Morganti, F Auricchio, DJ Benson, FI Gambarin, S Hartmann, TJR Hughes, and A Reali. Patient-specific isogeometric structural analysis of aortic valve closure. Computer Methods in Applied Mechanics and Engineering, 284:508 – 520, 2015 [6] Fei Xu, Simone Morganti, Rana Zakerzadeh, David Kamensky, Ferdinando Auricchio, Alessandro Reali, Thomas JR Hughes, Michael S Sacks, and Ming-Chen Hsu. A framework for designing patient-specific bioprosthetic heart valves using immersogeometric fluid – structure interaction analysis. International journal for numerical methods in biomedical engineering, 34(4):e2938, 2018. March 18, 2019 12

Valve Reconstruction and design • Interface valve with the patient’s aortic root • Define parameters for the designing • Vary them to get good performance March 18, 2019 13

Parametric Design of Heart Valve Geometry • Parameters of the heart valve affecting the geometry • Belly curvature ( x 3 ) • Height of free edge ( x 2 ) • Curvature of free edge ( x 1 ) [5] S Morganti, F Auricchio, DJ Benson, FI Gambarin, S Hartmann, TJR Hughes, and A Reali. Patient-specific isogeometric structural analysis of aortic valve closure. Computer Methods in Applied Mechanics and Engineering, 284:508 – 520, 2015 [6] Fei Xu, Simone Morganti, Rana Zakerzadeh, David Kamensky, Ferdinando Auricchio, Alessandro Reali, Thomas JR Hughes, Michael S Sacks, and Ming-Chen Hsu. A framework for designing patient-specific bioprosthetic heart valves using immersogeometric fluid – structure interaction analysis. International journal for numerical methods in biomedical engineering, 34(4):e2938, 2018. March 18, 2019 14

Non-Uniform Rational B-Spline Representation • Approximate the geometry using: • Control Points • Basis Functions (Piecewise polynomial) • Knot vectors • Weights A sample NURBS curve representation [7] http://web.me.iastate.edu/idealab/c-nurbs-python.html March 18, 2019 15

Non-Uniform Rational B-Spline Representation • Approximate the geometry using: • Control Points • Basis Functions • Knot vectors • Weights • Tensor Product for surfaces A sample NURBS surface representation [7] http://web.me.iastate.edu/idealab/c-nurbs-python.html March 18, 2019 16

Non-Uniform Rational B-Spline Representation De facto surface representation • Most general spline • Piecewise-polynomial tensor product surfaces • Can represent complex geometry such as heart valves [7] https://github.com/orbingol/NURBS-Python [8] Piegl, L., & Tiller, W. (2012). The NURBS book . Springer Science & Business Media. March 18, 2019 17

NURBS Evaluation Compact definition: z Defined completely by S ( 0,1 ) ( 0,1 ) ( 1,1 ) Control mesh  v 3 S ( 1,1 ) u and v knot vectors S ( u 0 , v 0 )  ( u 0 ,v 0 ) v v 2 S ( 0,0 ) v 1 y x u 1 u 2 u 3 ( 0,0 ) ( 0,1 ) S ( 1,0 ) u Parametric Model Space Space Knots ( u or v) Basis Functions Control Points p = degree   m n u u  u u       i p 1 p q p p 1 p 1 i N ( ) u N ( ) v w P N ( ) u N ( ) u N ( ) u    i i i 1 i j ij ij u u u u        j 0 i 0 i p i i p 1 i 1 S u v ( , )    m n  1 if u u u  p q   0 i i 1 N ( ) u N ( ) v w N ( ) u i j ij i  0 otherwise   j 0 i 0 March 18, 2019 18

Patient Specific Design of Heart Valve Geometry [9] https://web.me.iastate.edu/jmchsu/heart-valves.html March 18, 2019 19

Isogeometric Analysis • Based on technologies such as NURBS • Same (“exact”) functional description is used for geometry and simulation . • Includes standard FEA as a special case, but offers other possibilities:  Precise and efficient geometric modeling  Superior approximation properties  Smooth and higher-order basis functions  Integration of design and analysis March 18, 2019 20

CAD Model Quadratic NURBS Linear FEM Isogeometric Analysis FEM IGA Analysis Mesh CAD Model Coarse Mesh Refined Mesh

Challenges of Using IGA for BHV Design • Patient-specific design of bioprosthetic heart valves (BHV) require extensive exploration of design parameter space • Computational analysis is tedious and compute intensive • Lot of historical simulation data • Real-time decision support tool for analyzing valve function is difficult M.C. Hsu et.al., “Dynamic and fluid– structure interaction simulations of bioprosthetic heart valves using parametric design with T-splines and Fung- type material models,” Computational Mechanics, 55 (2015) 1211-1225 March 18, 2019 22

Deep Learning • Lots of Uses in Medical Sciences • Can learn complex phenomenon like the biomechanics • Can provide real-time support March 18, 2019 23

Outline Computational ML Framework for Data Generation Results and Modeling of BHVs Valve Mechanics Conclusions March 18, 2019 24

ML Framework for Valve Biomechanics Physical characteristics to learn: Embed phenomenon for learning: 1. Learn from 3D input space and predict 3D output deformation 1. Data Representation 2. Learn the effect of loads and boundary 2. Model Representation conditions 3. Training Algorithms 3. Interaction among the different leaflets 4. Material behavior and dependence on the thickness of the leaflets March 18, 2019 25